Database Schema Matching Using Machine Learning with Feature Selection

Berlin, Jacob; Motro, Amihai

doi:10.1007/3-540-47961-9_32

Cited by 64 publications

(11 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many research papers make use of principles from the database community to tackle machine learning problems, and vice versa (see for example [6,3,1]). Recently, relational learning has been proposed [25,20], an alternative machine learning approach based on declarative relational representations paired with probabilistic models.…”

Section: Data Dependencies For Machine Learningmentioning

confidence: 99%

Evaluating Classification Feasibility Using Functional Dependencies

Guilly

Petit

Scuturici

2020

Transactions on Large-Scale Data- And Knowledge-Centered Systems XLIV

View full text Add to dashboard Cite

With the vast amount of available tools and libraries for data science, it has never been easier to make use of classification algorithms: a few lines of code are enough to apply dozens of algorithms on any dataset. It is therefore "super easy" for data scientists to produce machine learning (ML) models in a very limited time. On the counterpart, domain experts may have the impression that such ML models are just a black box, almost magic, that would work on any dataset without really understanding why. For this reason, related to interpretability of machine learning, there is an urgent need to reconcile domain experts with ML models and to identify at the right level of abstraction, techniques to get them implied in the ML model construction.In this paper, we address this notion of trusting ML models by using data dependencies. We argue that functional dependencies characterize the existence of a function that a classification algorithm seeks to define. From this simple yet crucial remark, we have made several contributions. First, we show how functional dependencies can give a tight upper bound for the classification's accuracy, leading to impressive experimental results on UCI datasets with state-of-the art ML methods. Second, we point out how to generate very difficult synthetic datasets for classification, showing evidence about the fact that for some datasets, it does not make any sense to use ML methods. Third, we propose a practical and scalable solution to assess the existence of a function before applying ML techniques, allowing to take into account real life data and to keep domain experts in the loop.

show abstract

Section: Data Dependencies For Machine Learningmentioning

confidence: 99%

Evaluating Classification Feasibility Using Functional Dependencies

Guilly

Petit

Scuturici

2020

Transactions on Large-Scale Data- And Knowledge-Centered Systems XLIV

View full text Add to dashboard Cite

show abstract

“…Instance-based schema matching has been investigated by numerous studies that concentrate on enhancing the accuracy of the schema matching result [3,[6][7][12][13][14][15][16][17][18]. Different approaches have been proposed, adopted various strategies for precise determination of correspondence between attributes of schemas.…”

Section: Related Workmentioning

confidence: 99%

“…The work in [16] highlighted the issue of schema matching for a relational database. A machine learning strategy based approach named Autoplex is proposed to identify the match between schema attributes exploiting data instances.…”

Section: Related Workmentioning

confidence: 99%

An Empirical Comparative Study of Instance-based Schema Matching

Alzeber

Alwan

Nordin

et al. 2018

IJEECS

View full text Add to dashboard Cite

<span lang="EN-US">The main issue concern of schema matching is how to support the merging decision by providing matching between attributes of different schemas. There have been many works in the literature toward utilizing database instances to detect the correspondence between attributes. Most of these previous works aim at improving the match accuracy. We observed that no technique managed to provide an accurate matching for different types of data. In other words, some of the techniques treat numeric values as strings. Similarly, other techniques process textual instance, as numeric, and this negatively influences the process of discovering the match and compromising the matching result. Thus, a practical comparative study between syntactic and semantic techniques is needed. The study emphasizes on analyzing these techniques to determine the strengths and weaknesses of each technique. This paper aims at comparing two different instance-based matching techniques, namely: (i) regular expression and (ii) Google similarity to identify the match between attributes. Several analyses have been conducted on real and synthetic data sets to evaluate the performance of these techniques with respect to Precision (P), Recall (R) and F-Measure.</span>

show abstract

“…MaSiMe (Martinez-Gil & Aldana-Montes, 2011), Genetic Algorithm-based Ontology Matching (GAOM; Wang et al, 2006), Genetics for Ontology Alignments (GOAL; Martinez-Gilet al, 2008), eTuner (Lee et al, 2007), Alignment Process Feature Estimation and Learning (APFEL; Ehrig et al, 2005), MatchPlanner (Duchateau et al, 2008), and YAM (Yet Another Matcher; Duchateau et al, 2009) could be considered pure tools, while other tools are considered because they implement ontology meta-matching in any of the steps that they follow to solve problems. It should also be taken into account that several tools like Automatch (George Mason University; Berlin & Motro, 2002), GLUE (University of Washington; Doan et al, 2003), SemInt (C&C/MITRE Corporation/Oracle; Li & Clifton, 2000), and Rank Aggregation (Cornell University/Israel Institute of Technology; Domshlak et al, 2007) can only process classic schemas, and will therefore not be considered in this overview. The most outstanding tools in the area of heuristic ontology meta-matching are the following:…”

Section: Thresholdsmentioning

confidence: 99%

An overview of current ontology meta-matching solutions

Martinez‐Gil

Aldana‐Montes

2012

The Knowledge Engineering Review

View full text Add to dashboard Cite

Nowadays there are a lot of techniques and tools for addressing the ontology matching problem, however, the complex nature of this problem means that the existing solutions are unsatisfactory. This work intends to shed some light on a more flexible way of matching ontologies using ontology meta-matching. This emerging technique selects appropriate algorithms and their associated weights and thresholds in scenarios where accurate ontology matching is necessary. We think that an overview of the problem and an analysis of the existing state-of-the-art solutions will help researchers and practitioners to identify the most appropriate specific features and global strategies in order to build more accurate and dynamic systems following this paradigm.

show abstract

Database Schema Matching Using Machine Learning with Feature Selection

Cited by 64 publications

References 5 publications

Evaluating Classification Feasibility Using Functional Dependencies

Evaluating Classification Feasibility Using Functional Dependencies

An Empirical Comparative Study of Instance-based Schema Matching

An overview of current ontology meta-matching solutions

Contact Info

Product

Resources

About